Under the strictest interpretation, culpability for aiding and abetting would require an accomplice to fully embrace all elements of the principal offender’s crime, “positively to desire,” in James Stewart’s words, “the criminal outcome her assistance helps to bring into the world.”127 This view comports with how the Model Penal Code generally defines purposeful conduct on the part of a principal (“American criminal law has advanced significantly towards providing . . . precision, clarity, and rationality, owing in large part to the Model Penal Code. The common law and older codes often defined an offense to require only a single mental state. Under this ‘offense analysis,’ one spoke of intentional offenses, reckless offenses, and negligent offenses. The general culpability provisions of the Model Penal Code, in contrast, recognize that a single offense definition may require a different culpable state of mind for each objective element of the offense.”). Reflecting these ambiguities, English case law, for example, has divided on whether accomplices must actually desire to promote the criminal acts of principal perpetrators. See Duff, supra note 119, at 165–66 (observing that “[a]iding and abetting requires an ‘intention’ to assist the commission of the relevant offence” and surveying the split between, on the one hand, English cases holding that an accomplice has “that intention just so long as she acts in a way which she knows will facilitate the commission of the offence,” and, on the other hand, Scottish cases and some English cases, finding that intention requires more than mere knowledge); NEHA JAIN,PERPETRATORS AND ACCESSORIES IN INTERNATIONAL CRIMINAL LAW: INDIVIDUAL MODES OF RESPONSIBILITY FOR COLLECTIVE
CRIMES 161 (2014) (“There is controversy over whether [an accessory] must possess a
mental attitude with respect to the offence going beyond mere awareness and actually intend the offence to occur . . . [and] [t]he bulk of English authorities support the conclusion that D’s knowledge that his act will assist is sufficient.”). In the United States, by contrast, the majority position is that purpose, rather than knowledge, is required. See Paul H. Robinson, Imputed Criminal Liability, 93 YALE L.J. 609, 637 n.100 (1984) (“The defendant must have either a purpose to assist or encourage or, in some jurisdictions, simply have knowledge of the fact that he is assisting or encouraging. The majority of United States codes require purposeful or intentional assistance.”).
125. See Paul H. Robinson & Markus D. Dubber, The American Model Penal Code: A Brief Overview, 10 NEW CRIM.L.REV. 319, 335 (2007) (“In its zeal to clarify the law, the Model Penal Code even excised the words ‘intent’ and ‘intention’ from its terminology, concepts that in spite (or perhaps partly because) of their ambiguity had assumed a central place in the criminal law of the United States, as well as of many other countries.”).
126. See generally John F. Decker, The Mental State Requirement for Accomplice Liability in American Criminal Law, 60 S.C. L. REV. 237 (2008).
127. James G. Stewart, Complicity, in THE OXFORD HANDBOOK OF CRIMINAL LAW 534, 551 (Markus D. Dubber & Tatjana Hörnle eds., 2014) [hereinafter Stewart, Complicity].
2016] FOREIGN ASSISTANCE COMPLICITY 567
offender who “acts purposely with respect to a material element of an offense when[,] . . . if the element involves the nature of his conduct or a result thereof, it is his conscious object to engage in conduct of that nature or to cause such a result.”128
The obvious attraction of this approach lies in its protection against over-criminalization. Imagine, for example, that Perišić had disbursed arms to Bosnian Serb forces with precise instructions that tied the procurement of continued aid to the continued perpetration of atrocities. In such a case, the dilemmas of “specific direction” disap- pear. That Perišić might have known or even desired that the benefi- ciaries of his assistance would devote a portion of this support to le- gitimate combat functions is hardly relevant. By working to promote these atrocities, Perišić made them his own, reducing the need for further doctrinal protections against an overly expansive accomplice liability.
This is not to say, however, that a strict purpose requirement avoids all risk of over-criminalization. Consider the citizen that complies with her general legal duty to pay taxes in the knowledge that a portion of the revenue will end up supporting government cor- ruption. Imagine that the taxpayer happens to endorse this corruption and finds it to be a decisive reason to avoid tax evasion. Or imagine a variation on Perišić, in which a government official is responsible for overseeing an otherwise-permissible foreign aid program. The official knows that a small portion of the assistance will find its way to recipients who perpetrate atrocities, and the official has volun- teered for this work precisely because he wants to support these atrocities. Nevertheless, the official undertakes his work in a neutral matter, performing his duties exactly as he might if he did not em- brace this purpose. As these admittedly fanciful examples illustrate, even a strict mens rea requirement may be insufficient—without oth- er legal protections—to ground criminal responsibility in cases of minor or attenuated contributions to criminal activity.
Nevertheless, the far more substantial problem with this strict approach is that it is under-inclusive, leading to well-known arbitrary distinctions between accomplices and principal perpetrators. Stewart gives the example of a person who, acting without justification, throws a grenade into a passing car, knowing to a virtual certainty that the explosion will kill someone inside.129 The grenade explosion 128. MODEL PENAL CODE § 2.02(2)(a)–(a)(i) (AM.LAW INST. 1962) (emphasis added). The Code further provides that the offender acts purposely with respect to background facts, or so-called “attendant circumstances,” if “he is aware of their existence or he believes or hopes that they exist.” MODEL PENAL CODE § 2.02(2)(a)(ii) (AM.LAW INST. 1962).
568 COLUMBIA JOURNAL OF TRANSNATIONAL LAW [54:531
does indeed kill the car’s occupants and the perpetrator (whom I will name Sid) is guilty of murder.130 Whether or not Sid acted with the express purpose to kill is irrelevant. It might be, for example, that the car’s driver has just taken an embarrassing photograph that Sid wants to destroy, and that it is all the same to him whether or not any of the car’s occupants survive the explosion. This non-homicidal purpose fails to exonerate in light of Sid’s awareness that his actions were vir- tually certain to result in death. Indeed, even something less than vir- tual certainty will be sufficient to support a homicide conviction—if not for murder, then for a lower grade of homicide.
Suppose, now, that Sid is assisted by an accomplice, Nancy, who passes him the grenade after pulling the pin. Suppose, moreo- ver, that Nancy knows full well what Sid intends to do, but is person- ally indifferent regarding whether or not Sid plans to throw the gre- nade into the passing vehicle or, instead, into a nearby pond where it will safely explode without risk to human life. What reason is there, in a case like this, to convict Sid of murder on the ground that he was aware of the certain consequences of his action, but to absolve his accomplice, Nancy, on the ground that her identical knowledge does not amount to a purpose to take human life?
Different versions of this scenario—both hypothetical and re- al-life—have played a prominent role in scholarly and judicial criti- cism of the strictly purposive approach.131 Notably, this scenario does not benefit from the rationale commonly invoked to defend a purpose requirement. Handing a live grenade to a would-be murder- 130. This result follows from the common law requirement of malice aforethought, see, for example, Regina v. Serné, 16 Cox. Crim. Cas. 311 (1887), reprinted in AUGUSTIN
DERBY, CASES ON CRIMINAL LAW 342 (1917) (“The definition [of murder] is unlawful
homicide with malice aforethought; and the words malice aforethought are technical. . . . [One] meaning is, an act done with the knowledge that the act will probably cause the death of some person.”) and from the Model Penal Code’s approach to homicide, see MODEL
PENAL CODE § 210.2 (AM.LAW INST. 1962) (providing, inter alia, that “criminal homicide
constitutes murder when . . . it is committed purposely or knowingly”).
131. Glanville Williams, for instance, cites a similar example in support of the proposition that the mens rea required to prosecute aiders and abettors may sometimes be looser than for principal perpetrators. See WILLIAMS, supra note 122, at 394 (“It seems that even where a crime requires intention on the part of the perpetrator, intention need not be proved against a secondary party if he knowingly involves himself in the affair. Thus: D is about to kill P, and requests E to hand him a weapon with which to do it; E hands him the weapon, not because he wishes P to die, but because he always likes to oblige D. E is clearly a party to the murder if he knows of D’s intention. In this respect, the mental element required for secondary parties is more widely defined than the mental element required for the perpetrator.”); see, e.g., United States v. Fountain, 768 F.2d 790 (7th Cir. 1985) (rejecting the purposive approach to complicity in the case of an inmate who provided the knife to a fellow inmate, which was used to murder a guard).
2016] FOREIGN ASSISTANCE COMPLICITY 569
er has little to do with selling everyday products in the ordinary course of business. Nor is it intuitive to describe culpability of such an accomplice as a form of guilt by omission. Accordingly, to the extent that the “ordinary conduct of business” argument provides a persuasive reason to reject some kinds of knowledge-based complici- ty, this argument does not cover all cases.